与之前的版本校对

UpdateCi.py

@@ -0,0 +1,136 @@
+import json
+import logging
+import os
+import re
+from difflib import SequenceMatcher
+
+import requests
+from bs4 import BeautifulSoup
+from bs4.element import NavigableString
+
+
+def get_page_content(page: int) -> list:
+    """ 获取目录页每一页的内容 """
+    content = []
+    r = requests.post("http://qsc.zww.cn/getdata.asp", data={
+        "seektype": 2,
+        "seekvalue": "",
+        "pageno": page
+    })
+    r.encoding = "gbk"
+    soup = BeautifulSoup(re.search(r"filllist\('·(.*?)'\);", r.text).group(1), features="lxml")
+    for i, a in enumerate(soup.find_all(name="a")):
+        if i % 2 == 0:
+            content.append({
+                "rhythmic": a.string.split("（")[0],
+                "param": re.search(r"doseek2\((.*?)\);", a["onclick"]).group(1).split(",")
+            })
+        else:
+            content[-1]["author"] = a.string
+    for c in content:
+        c["paragraphs"] = get_paragraphs(int(c["param"][0]), int(c["param"][1]))
+        del c["param"]
+    return content
+
+
+def get_paragraphs(seek_type: int, seek_value: int) -> list:
+    """ 获取词的内容段落 """
+    paragraphs = []
+    r = requests.post("http://qsc.zww.cn/getdata.asp", data={
+        "seektype": seek_type,
+        "seekvalue": seek_value,
+        "pageno": 1
+    })
+    r.encoding = "gbk"
+    soup = BeautifulSoup(re.search(r"fillbody\('(.*?)'\);", r.text).group(1), features="lxml")
+    for child in soup.find(name="p", align=None).contents:
+        if isinstance(child, NavigableString):
+            paragraphs.append(child)
+    return paragraphs
+
+
+def get_all_page(temp_file: str):
+    """ 爬取数据并保存至临时文件 """
+    for page in range(1, 1240):
+        all_data.extend(get_page_content(page))
+        logging.info("Success: save page {0}".format(page))
+    with open(temp_file, "w", encoding="utf-8") as f:
+        f.write(json.dumps(all_data, indent=2, ensure_ascii=False))
+
+
+def only_text(text: str):
+    """ 去除标点只保留文字 """
+    return re.sub(r"[，。、《》…（）·・\s]", "", text)
+
+
+def update_file_data(old_data: list, new_data: list):
+    for i in range(len(old_data)):
+        old_text = only_text("".join(old_data[i]["paragraphs"]))
+        new_text = only_text("".join(new_data[start + i]["paragraphs"]))
+        # 计算纯文字的相似度
+        ratio = SequenceMatcher(a=old_text, b=new_text).quick_ratio()
+        if 0.9 <= ratio < 1.0:
+            # 假定此范围内说明缺字，需要更新
+            old_data[i]["author"] = new_data[start + i]["author"]
+            old_data[i]["paragraphs"] = new_data[start + i]["paragraphs"]
+        elif ratio < 0.9:
+            # 异常情况warning输出，不更新
+            logging.warning(old_text)
+            logging.warning(new_text)
+        else:
+            old_data[i]["author"] = new_data[start + i]["author"]
+
+
+char_dict = {
+    "鵷": "鹓",
+    "颭": "飐",
+    "鷁": "鹢",
+    "鴞": "鸮",
+    "餖": "饾",
+    "飣": "饤",
+    "舃": "舄",
+    "駸": "骎",
+    "薄倖": "薄幸",
+    "赬": "赪",
+    "鷫鸘": "鹔鹴",
+    "嶮": "崄",
+    "後": "后",
+    "纇": "颣",
+    "颸": "飔",
+    "崑崙": "昆仑",
+    "曨": "昽"
+}
+
+
+def correct(old_data: list):
+    """ 部分繁体转为简体 """
+    for i in range(len(old_data)):
+        for j in range(len(old_data[i]["paragraphs"])):
+            for k, v in char_dict.items():
+                if k in old_data[i]["paragraphs"][j]:
+                    old_data[i]["paragraphs"][j] = old_data[i]["paragraphs"][j].replace(k, v)
+
+
+if __name__ == '__main__':
+    logging.basicConfig(level=logging.INFO,
+                        format="%(asctime)s - %(levelname)-9s %(filename)-15s[:%(lineno)d]\t%(message)s")
+    temp_file_name = "all.json"
+    # 临时文件不存在则先爬取
+    if not os.path.exists(temp_file_name):
+        get_all_page(temp_file_name)
+    # 读取临时文件
+    with open("all.json", "r", encoding="utf-8") as f:
+        all_data = json.load(f)
+    # 遍历当前目录
+    for file_name in os.listdir("./ci/"):
+        if re.match(r"ci\.song\.\d+\.json", file_name):
+            # 每个文件开始的数据索引
+            start = int(file_name.split(".")[2])
+            with open("./ci/" + file_name, "r", encoding="utf-8") as f:
+                file_data = json.load(f)
+            # update_file_data(file_data, all_data)
+            correct(file_data)
+            # 保存数据，原文件中逗号后有空格，这里保持一致
+            with open("./ci/" + file_name, "w", encoding="utf-8") as f:
+                f.write(json.dumps(file_data, indent=2, ensure_ascii=False).replace(",", ", "))
+                logging.info("Save " + file_name)

ci/UpdateCi.py

@@ -63,6 +63,54 @@ def only_text(text: str):
     return re.sub(r"[，。、《》…（）·・\s]", "", text)
 
 
+def update_file_data(old_data: list, new_data: list):
+    for i in range(len(old_data)):
+        old_text = only_text("".join(old_data[i]["paragraphs"]))
+        new_text = only_text("".join(new_data[start + i]["paragraphs"]))
+        # 计算纯文字的相似度
+        ratio = SequenceMatcher(a=old_text, b=new_text).quick_ratio()
+        if 0.9 <= ratio < 1.0:
+            # 假定此范围内说明缺字，需要更新
+            old_data[i]["author"] = new_data[start + i]["author"]
+            old_data[i]["paragraphs"] = new_data[start + i]["paragraphs"]
+        elif ratio < 0.9:
+            # 异常情况warning输出，不更新
+            logging.warning(old_text)
+            logging.warning(new_text)
+        else:
+            old_data[i]["author"] = new_data[start + i]["author"]
+
+
+char_dict = {
+    "鵷": "鹓",
+    "颭": "飐",
+    "鷁": "鹢",
+    "鴞": "鸮",
+    "餖": "饾",
+    "飣": "饤",
+    "舃": "舄",
+    "駸": "骎",
+    "薄倖": "薄幸",
+    "赬": "赪",
+    "鷫鸘": "鹔鹴",
+    "嶮": "崄",
+    "後": "后",
+    "纇": "颣",
+    "颸": "飔",
+    "崑崙": "昆仑",
+    "曨": "昽"
+}
+
+
+def correct(old_data: list):
+    """ 部分繁体转为简体 """
+    for i in range(len(old_data)):
+        for j in range(len(old_data[i]["paragraphs"])):
+            for k, v in char_dict.items():
+                if k in old_data[i]["paragraphs"][j]:
+                    old_data[i]["paragraphs"][j] = old_data[i]["paragraphs"][j].replace(k, v)
+
+
 if __name__ == '__main__':
     logging.basicConfig(level=logging.INFO,
                         format="%(asctime)s - %(levelname)-9s %(filename)-15s[:%(lineno)d]\t%(message)s")
@@ -73,8 +121,6 @@ if __name__ == '__main__':
     # 读取临时文件
     with open("all.json", "r", encoding="utf-8") as f:
         all_data = json.load(f)
-    # 统计更正的数目
-    diff_num = 0
     # 遍历当前目录
     for file_name in os.listdir("./"):
         if re.match(r"ci\.song\.\d+\.json", file_name):
@@ -82,22 +128,9 @@ if __name__ == '__main__':
             start = int(file_name.split(".")[2])
             with open(file_name, "r", encoding="utf-8") as f:
                 file_data = json.load(f)
-            for i in range(len(file_data)):
-                old_text = only_text("".join(file_data[i]["paragraphs"]))
-                new_text = only_text("".join(all_data[start + i]["paragraphs"]))
-                # 计算纯文字的相似度
-                ratio = SequenceMatcher(a=old_text, b=new_text).quick_ratio()
-                if 0.9 <= ratio < 1.0:
-                    # 假定此范围内说明缺字，需要更新
-                    diff_num += 1
-                    file_data[i]["author"] = all_data[start + i]["author"]
-                    file_data[i]["paragraphs"] = all_data[start + i]["paragraphs"]
-                elif ratio < 0.9:
-                    # 异常情况warning输出，不更新
-                    logging.warning(old_text)
-                    logging.warning(new_text)
+            update_file_data(file_data, all_data)
+            correct(file_data)
             # 保存数据，原文件中逗号后有空格，这里保持一致
             with open(file_name, "w", encoding="utf-8") as f:
                 f.write(json.dumps(file_data, indent=2, ensure_ascii=False).replace(",", ", "))
                 logging.info("Save " + file_name)
-    logging.info("Change {0} items".format(diff_num))

ci/ci.song.21000.json

@@ -146,7 +146,7 @@
       "幕天席地。", 
       "瑞脑香浓笙歌沸。", 
       "白纻衣轻。", 
-      "鵷发霜髯照座明。", 
+      "鹓发霜髯照座明。", 
       "轻簪小珥。", 
       "却是人间真富贵。", 
       "好着丹青。", 
@@ -214,7 +214,7 @@
     "paragraphs": [
       "香烟绕遍兰堂宴。", 
       "香鸭珠帘卷。", 
-      "香风转後送韶音。", 
+      "香风转后送韶音。", 
       "香酝佳筵今日、庆佳辰。", 
       "香山烧尽禽飞放。", 
       "香袖佳人唱。", 
@@ -327,7 +327,7 @@
       "昴宿呈祥，符应生公族。", 
       "盖世功名夸九牧。", 
       "黼衮褒扬，庆阀辉南北。", 
-      "赐宫醪，分笃<EFBFBD>酢<EFBFBD>", 
+      "赐宫醪，分笃耨。", 
       "天与长生，谩把仙椿祝。", 
       "好继平阳腾茂躅。", 
       "富贵千秋，饮听瑶池曲。"
@@ -565,7 +565,7 @@
       "龙角辉春，蛾春惊晓，梦阑金翠屏开。", 
       "异芬薰室，风送蕊仙来。", 
       "玉女擎香沐浴，人间世、洗彻凡埃。", 
-      "梅开後，留花酝染，清味俗难猜。", 
+      "梅开后，留花酝染，清味俗难猜。", 
       "东君，尤雅爱，传香芳畹，香发庭陔。", 
       "宁馨满尊前，喜奏瑶台。", 
       "便好纽为佩王，瀛洲路、同赏蓬莱。", 
@@ -576,7 +576,7 @@
   {
     "author": "潘熊飞", 
     "paragraphs": [
-      "十日後重阳。", 
+      "十日后重阳。", 
       "甘菊阶前满意黄。", 
       "生日无钱留贺客，何妨。", 
       "尚有儿曹理寿觞。",